RENDITA Project: Launch of Advanced Digital Platforms for Multi-Generation Energy System Management

The RENDITA (REsilient Network through DIgital Twin Applications) project has officially started, funded with over €1 million under the Mission Innovation 2.0 program – “Data and Grid Digitalization”.
The project, coordinated by 3rd Place S.r.l., is part of the transition towards Multi-Generation Energy Systems (MGS), integrated systems that combine different energy networks – including electricity, gas, and district heating – with the aim of improving efficiency, resilience, and sustainability. Coordinated management of these infrastructures, together with the ability to predict failures and optimize operations, represents one of the main current technological challenges.
RENDITA aims to develop a next-generation platform based on Digital Twin (DT) and AI to support Operation & Maintenance (O&M) activities in complex energy systems.

The project involves the Department of Energy of the Politecnico di Milano, which is responsible for the development of two key Work Packages:

  • WP4 – Grey-Box Models (GBMs): hybrid models that combine physical system knowledge and machine learning techniques, enabling advanced Digital Twin functionalities;
  • WP7 – Multi-Agent Reinforcement Learning (MARL): multi-agent reinforcement learning approaches applied to the optimization of operational and maintenance strategies in multi-energy systems.

The RENDITA approach also integrates Prognostics & Health Management (PHM) methodologies, which allow to:

  • estimate the health status of assets,
  • early identify potential failures,
  • improve alignment between system performance, weather conditions, and energy demand.

Through the combined use of AI, IoT, and DT technologies, the project aims to improve the reliability and efficiency of energy infrastructures, contributing to the development of smarter and more resilient energy networks.

 
 
“RENDITA represents an important step towards smarter and more predictive management of complex energy systems,” comments Enrico Zio, professor in the Department of Energy at the Politecnico di Milano and head of the Department’s research activities in the project. “The integration of hybrid models, Digital Twin technologies, and advanced learning techniques allows us to overcome the limitations of traditional approaches, enabling more efficient, resilient, and data-driven operation and maintenance strategies.”